From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jiri Pirko <jiri@resnulli.us>
Subject: Re: Fwd: u32 ht filters
Date: Mon, 12 Feb 2018 16:32:16 +0100
Message-ID: <20180212153216.GB2153@nanopsycho>
References: <9c8f997d-339c-5088-0bb5-124e9f55f02d@itcare.pl>
 <CAM_iQpX_CG9casEeFiXtnBNUg15ZWrdskp3JbFnXd=zXBeb7cg@mail.gmail.com>
 <20180207070148.GA2149@nanopsycho.orion>
 <CAM_iQpXkmSPGXfbOyLH-BMT+_u4MfEG+y78H5JNr9x1i=nNTJw@mail.gmail.com>
 <20180208073846.GA2041@nanopsycho>
 <CAM_iQpXwKmuuDaHWuJ+E60P2pr00tsVr2+p_hY++D7Vi8Hbavw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Jiri Pirko <jiri@mellanox.com>,
        Linux Kernel Network Developers <netdev@vger.kernel.org>,
        =?utf-8?B?UGF3ZcWC?= Staszewski <pstaszewski@itcare.pl>
To: Cong Wang <xiyou.wangcong@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-wm0-f50.google.com ([74.125.82.50]:51177 "EHLO
        mail-wm0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752281AbeBLPcT (ORCPT
        <rfc822;netdev@vger.kernel.org>); Mon, 12 Feb 2018 10:32:19 -0500
Received: by mail-wm0-f50.google.com with SMTP id f71so10328780wmf.0
        for <netdev@vger.kernel.org>; Mon, 12 Feb 2018 07:32:18 -0800 (PST)
Content-Disposition: inline
In-Reply-To: <CAM_iQpXwKmuuDaHWuJ+E60P2pr00tsVr2+p_hY++D7Vi8Hbavw@mail.gmail.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

Sat, Feb 10, 2018 at 09:41:57PM CET, xiyou.wangcong@gmail.com wrote:

[...]

>>>@@ -348,7 +340,7 @@ static struct tc_u_common *tc_u_common_find(const
>>>struct tcf_proto *tp)
>>>
>>>        h = tc_u_hash(tp);
>>>        hlist_for_each_entry(tc, &tc_u_common_hash[h], hnode) {
>>>-               if (tc->block == tp->chain->block)
>>>+               if (tc->block->q == tp->chain->block->q)
>>
>> :O I don't get it. tc->block is pointer, tc->block->q is pointer. And
>> they are different at the same time for non-shared block.
>
>If you look into Pawel's script, a new block is created for each class
>therefore a different tc_u_common is created which causes the
>ht 9:22 can't be found.

But wait. Originally this code looked like this:
static unsigned int tc_u_hash(const struct tcf_proto *tp)
{
        struct net_device *dev = tp->q->dev_queue->dev;
        u32 qhandle = tp->q->handle;
        int ifindex = dev->ifindex;

        return hash_64((u64)ifindex << 32 | qhandle, U32_HASH_SHIFT);
}

static struct tc_u_common *tc_u_common_find(const struct tcf_proto *tp)
{
        struct tc_u_common *tc;
        unsigned int h;

        h = tc_u_hash(tp);
        hlist_for_each_entry(tc, &tc_u_common_hash[h], hnode) {
                if (tc->q == tp->q)
                        return tc;
        }
        return NULL;
}

That means that tc_u_common is identified according to tp->q. And that
is different for every class. How that could work? I'm properly confused.